SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.14628 · 2020

Solving the Blind Perspective-n-Point Problem End-To-End With Robust Differentiable Geometric Optimization

arXiv · PDF · Open in the Atlas

Code that ran

18 of 19 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 9 have no contract verdict; 8 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2007.14628.json · badge: /api/ran/2007.14628.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
Liumouliu/Deep_blind_PnP canonical (classifier label)
classifier basis: plan_row
17 ran · 1 no recorded run
dylan-campbell/bpnpnet canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
Dy Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
dylan-campbell/bpnpnet/lib/nonlinear_weighted_blind_pnp.py
code served (permissive licence) · get_code("50906fc42db82ac9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 3, dtype family float, shape (2, 3, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Set up a simple test case for the Dy function
# The function computes implicit differentiation using the formula from the paper

# Create batch size and dimensions
batch_size = 2
n_x = 4  # dimension of x
m_y = 3  # dimension of y

# Create a simple quadratic function f(x, y) = 0.5 * y^T * A * y + x^T * B * y
# where A and B are learnable parameters
class SimpleObjective(nn.Module):
    def __init__(self):
        super().__init__()
        self.A = nn.Parameter(torch.randn(m_y, m_y))
        self.B = nn.Parameter(torch.randn(n_x, m_y))
        # Make A positive definite
        self.A.data = self.A.data @ self.A.data.T + torch.eye(m_y) * 0.5
    
    def forward(self, x, y):
        # f(x, y) = 0.5 * y^T * A * y + x^T * B * y
        quad_term = 0.5 * torch.sum(y @ self.A * y, dim=-1)
        cross_term = torch.sum(x @ self.B * y, dim=-1)
        return quad_term + cross_term

# Instantiate the objective function
objective_fn = SimpleObjective()

# Create test tensors
x = torch.randn(batch_size, n_x, requires_grad=True, dtype=torch.float32)
y = torch.randn(batch_size, m_y, requires_grad=True, dtype=torch.float32)

# Evaluate the objective function at (x, y)
f = objective_fn(x, y)

ARGS = {
    'f': f,
    'x': x,
    'y': y,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 4) — shape (2, 3, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
add_argument_group Ran
ran; our contract draft was wrong, not the code
Liumouliu/Deep_blind_PnP/config.py
code served (permissive licence) · get_code("6d80592fd6d47b2d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned _ArgumentGroup. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import argparse

arg_lists = []
parser = argparse.ArgumentParser()

def add_argument_group(name):
    arg = parser.add_argument_group(name)
    arg_lists.append(arg)
    return arg

ARGS = {
    'name': 'test_group'
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type _ArgumentGroup — a _ArgumentGroup output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a _ArgumentGroup output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
bn_act Ran
ran on a synthesized input
Liumouliu/Deep_blind_PnP/model/ops.py
code served (permissive licence) · get_code("5a0fa80f8d2294d9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Define ARGS dict with all parameters for bn_act function
ARGS = {
    'in_channel': 4,
    'perform_gcn': True,
    'perform_bn': True,
    'activation': 'relu'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
correspondenceLoss Ran
ran on a synthesized input
Liumouliu/Deep_blind_PnP/lib/loss.py
code served (permissive licence) · get_code("c53032885d220555")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn.functional as F

torch.manual_seed(0)

# P is a probability tensor used with log() and arithmetic operations
# C_gt is a correspondence/confidence tensor used with sum, arithmetic operations
# Both should have shape allowing sum over last 2 dimensions
# Batch size 2, spatial dimensions 3x4

P = torch.rand(2, 3, 4)  # Predicted probabilities, values in [0, 1]
C_gt = torch.randint(0, 2, (2, 3, 4)).float()  # Ground truth correspondences, binary values

ARGS = {
    "P": P,
    "C_gt": C_gt,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
correspondenceProbabilityBCE Ran
ran on a synthesized input
Liumouliu/Deep_blind_PnP/lib/loss.py
code served (permissive licence) · get_code("15fcc76407a58c98")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn.functional as F

torch.manual_seed(0)

# P: predicted probability tensor - should have shape (batch, height, width)
# C: correspondence/ground truth tensor - should have shape (batch, height, width)
# Both tensors should contain values in [0, 1] range for probability/binary labels

batch_size = 2
height = 4
width = 3

# Create P: predicted probabilities, shape (2, 4, 3)
P = torch.rand(batch_size, height, width)

# Create C: ground truth correspondence labels (0 or 1), shape (2, 4, 3)
C = torch.randint(0, 2, (batch_size, height, width)).float()

ARGS = {
    "P": P,
    "C": C,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
correspondenceProbabilityDistances Ran
ran on a synthesized input
Liumouliu/Deep_blind_PnP/lib/loss.py
code served (permissive licence) · get_code("3753d4559e502380")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# P: probability tensor, needs to support .sum(dim=(-2, -1))
# Shape: (batch, ..., height, width) where we sum over last two dimensions
P = torch.rand(2, 4, 3, 5)  # batch=2, intermediate=4, height=3, width=5

# C: correspondence mask/weights, same shape as P for broadcasting
# Values should be between 0 and 1 (probability-like)
C = torch.rand(2, 4, 3, 5)

ARGS = {
    "P": P,
    "C": C,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_file_list Ran
ran; our contract draft was wrong, not the code
Liumouliu/Deep_blind_PnP/lib/file.py
code served (permissive licence) · get_code("262f74c5ade997a1")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tempfile
import os

# Create a temporary directory with some test files
temp_dir = tempfile.mkdtemp()

# Create some test files with alphanumeric names
test_files = [
    "file10.txt",
    "file2.txt",
    "file1.txt",
    "document20.pdf",
    "document3.pdf",
    "image100.jpg",
]

for filename in test_files:
    filepath = os.path.join(temp_dir, filename)
    with open(filepath, 'w') as f:
        f.write("test content")

# Define arguments for get_file_list
ARGS = {
    "path": temp_dir,
    "extension": ".txt"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_graph_feature Ran
ran on a synthesized input
Liumouliu/Deep_blind_PnP/model/ops.py
code served (permissive licence) · get_code("f5dcd38523dc6217")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (2, 8, 3, 2) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function get_graph_feature expects:
# - x: a tensor with shape (batch_size, num_dims, num_points)
# - k: an integer for number of neighbors
# - idx: optional pre-computed neighbor indices

batch_size = 2
num_dims = 4
num_points = 3
k = 2

# Create a small input tensor with shape (batch_size, num_dims, num_points)
x = torch.randn(batch_size, num_dims, num_points)

ARGS = {
    "x": x,
    "k": k,
    "idx": None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 8, 3, 2) — shape (2, 8, 3, 2) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
knn Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Liumouliu/Deep_blind_PnP/model/ops.py
code served (permissive licence) · get_code("cdd0141594039dcb")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 3, dtype family int, shape (2, 3, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# x is used with:
# - transpose(2, 1): expects 3D tensor
# - matmul with transposed version: matrix multiplication
# - sum(x ** 2, dim=1, keepdim=True): sum over dimension 1
# - topk(k=k, dim=-1): get top k indices along last dimension
# 
# Shape should be (batch, features, num_points) based on the transpose operations
# Let's use batch=2, features=4, num_points=3

x = torch.randn(2, 4, 3)

# k is the number of nearest neighbors to find
# Should be less than num_points (3), so k=2
k = 2

ARGS = {
    "x": x,
    "k": k,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family int, shape (2, 3, 2) — shape (2, 3, 2) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_data_3d2d_pairs Ran
ran on a synthesized input
Liumouliu/Deep_blind_PnP/lib/data_loaders.py
code served (permissive licence) · get_code("bd6343637c69e84b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import pickle
import sys
import tempfile
import shutil

# Create a temporary directory structure with pickle files
temp_dir = tempfile.mkdtemp()

# Create dataset subdirectory
dataset_name = 'test_dataset'
dataset_split = 'train'
dataset_dir = os.path.join(temp_dir, f'{dataset_name}_{dataset_split}')
os.makedirs(dataset_dir, exist_ok=True)

# Create mock pickle files for the required variables
var_name_list = ['matches', 'focal', 'R', 't', 'inlier_matches']

# Create sample data for each variable
sample_data = {
    'matches': [[1, 2], [3, 4]],
    'focal': [500.0, 510.0],
    'R': [[1.0, 0.0, 0.0], [0.0, 1.0, 0.0], [0.0, 0.0, 1.0]],
    't': [0.1, 0.2, 0.3],
    'inlier_matches': [[1, 2], [3, 4], [5, 6]]
}

for var_name in var_name_list:
    file_path = os.path.join(dataset_dir, f'{var_name}.pkl')
    with open(file_path, 'wb') as f:
        pickle.dump(sample_data[var_name], f)

# Create a minimal config class
class Config:
    def __init__(self):
        self.data_dir = temp_dir
        self.dataset = dataset_name

config = Config()

ARGS = {
    'config': config,
    'dataset_split': dataset_split
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
make_data_loader Ran
ran on a synthesized input
Liumouliu/Deep_blind_PnP/lib/data_loaders.py
code served (permissive licence) · get_code("ad9e933f883a8298")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned DataLoader. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import torch
import torch.utils.data
import os
import pickle
import tempfile
import shutil

torch.manual_seed(0)

# Create a minimal config class
class Config:
    def __init__(self):
        self.dataset = 'test_dataset'
        self.data_dir = None  # Will be set to temp dir
        self.numPointsTrain = 8

# Create temporary directory with mock pickle files
temp_dir = tempfile.mkdtemp()
dataset_split = 'train'
dataset_folder = os.path.join(temp_dir, f'test_dataset_{dataset_split}')
os.makedirs(dataset_folder, exist_ok=True)

# Create mock data files
mock_data = {
    'matches': [np.random.rand(10, 5).astype(np.float32) for _ in range(2)],
    'focal': [np.array([500.0], dtype=np.float32) for _ in range(2)],
    'R': [np.eye(3, dtype=np.float32) for _ in range(2)],
    't': [np.array([0.1, 0.2, 0.3], dtype=np.float32) for _ in range(2)],
    'inlier_matches': [np.array([[0, 1], [2, 3], [4, 5]], dtype=np.int32) for _ in range(2)]
}

for var_name, data_list in mock_data.items():
    file_path = os.path.join(dataset_folder, f'{var_name}.pkl')
    with open(file_path, 'wb') as f:
        pickle.dump(data_list, f)

# Create config
config = Config()
config.data_dir = temp_dir

# Create ARGS dictionary
ARGS = {
    'config': config,
    'phase': 'train',
    'batch_size': 2,
    'num_threads': 0,
    'shuffle': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataLoader — a DataLoader output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataLoader output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
pairwiseL2Dist Ran
ran; our contract draft was wrong, not the code
Liumouliu/Deep_blind_PnP/model/model.py
code served (permissive licence) · get_code("c2f2bf68ef50b018")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 3, dtype family float, shape (2, 3, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create two small feature vector batches
# x1: shape (batch_size=2, num_vectors=3, feature_dim=4)
# x2: shape (batch_size=2, num_vectors=5, feature_dim=4)
x1 = torch.randn(2, 3, 4)
x2 = torch.randn(2, 5, 4)

ARGS = {
    "x1": x1,
    "x2": x2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 5) — shape (2, 3, 5) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
read_txt Ran
ran; our contract draft was wrong, not the code
Liumouliu/Deep_blind_PnP/lib/file.py
code served (permissive licence) · get_code("6f1f2152e78c5e6f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tempfile
import os

# Create a temporary text file with some content
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.txt")

with open(temp_file, 'w') as f:
    f.write("line one\n")
    f.write("line two\n")
    f.write("line three\n")

ARGS = {
    "path": temp_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
reflection_matrix Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Liumouliu/Deep_blind_PnP/lib/transformations.py
code served (permissive licence) · get_code("f7f6bfdeee237f86")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (4, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy
import math

# Construct arguments for reflection_matrix function
# The function takes:
# - point: should be a vector of length 4 (used as point[:3] for the plane point)
# - normal: should be a vector of length 3+ (used as normal[:3])

point = numpy.array([1.0, 2.0, 3.0, 1.0])
normal = numpy.array([0.5, 0.3, 0.2])

ARGS = {
    "point": point,
    "normal": normal,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (4, 4) — shape (4, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
sorted_alphanum Ran
ran; our contract draft was wrong, not the code
Liumouliu/Deep_blind_PnP/lib/file.py
code served (permissive licence) · get_code("03faf5df924907dd")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def sorted_alphanum(file_list_ordered):
    def convert(text):
        return int(text) if text.isdigit() else text

    def alphanum_key(key):
        return [convert(c) for c in re.split('([0-9]+)', key)]
    return sorted(file_list_ordered, key=alphanum_key)

# Setup code
file_list = [
    "file10.txt",
    "file2.txt",
    "file1.txt",
    "image20.png",
    "image3.png",
    "test100_data.csv",
    "test20_data.csv"
]

ARGS = {
    "file_list_ordered": file_list
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
str2bool Ran
ran, violated the contract we drafted
Liumouliu/Deep_blind_PnP/config.py
pointer only (licence: MIT); the stored copy get_code serves, from haoran0301/amenunet, is pointer only (licence: NONE) · get_code("248284f69adfeaad")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from haoran0301/amenunet, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

translation_from_matrix Ran
ran on a synthesized input
Liumouliu/Deep_blind_PnP/lib/transformations.py
code served (permissive licence) · get_code("77cb383b0cd2a54e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family float, shape (3,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy

def translation_matrix(direction):
    """Helper function to create a translation matrix from a direction vector."""
    matrix = numpy.eye(4)
    matrix[:3, 3] = direction
    return matrix

# Create a small translation vector
v0 = numpy.array([0.1, 0.2, 0.3])

# Create a translation matrix (4x4 homogeneous transformation matrix)
matrix = translation_matrix(v0)

ARGS = {
    "matrix": matrix
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (3,) — shape (3,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
translation_matrix Ran
ran on a synthesized input
Liumouliu/Deep_blind_PnP/lib/transformations.py
code served (permissive licence) · get_code("edeb1406911ce025")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (4, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy

numpy.random.seed(0)

# The function expects a direction vector with at least 3 elements
# It accesses direction[:3] to extract the first 3 components
direction = numpy.random.random(3) - 0.5

ARGS = {
    "direction": direction
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (4, 4) — shape (4, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_model Not yet run
unverified (harvested, no recorded run)
Liumouliu/Deep_blind_PnP/lib/utils.py
code served (permissive licence) · get_code("cf0cddae7aa96a4b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 19 distinct functions with an identity, 12 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 0 the same month, 12 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
knn 52 other papers
4 dated earlier, 48 later · 12 only as an assembly brick
1801.07829 (2018-01)
Dynamic Graph CNN for Learning on Point Clouds
1905.06292 (2019-05)
3D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutions
2005.06734 (2020-05)
Dense-Resolution Network for Point Cloud Classification and Segmentation
and 49 more not named here
str2bool 33 other papers
19 dated earlier, 14 later
1406.6247 (2014-06)
Recurrent Models of Visual Attention
1506.03134 (2015-06)
Pointer Networks
1611.09940 (2016-11)
Neural Combinatorial Optimization with Reinforcement Learning
and 30 more not named here
add_argument_group 27 other papers
16 dated earlier, 11 later
1406.6247 (2014-06)
Recurrent Models of Visual Attention
1506.03134 (2015-06)
Pointer Networks
1609.04802 (2016-09)
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
and 24 more not named here
reflection_matrix 14 other papers
8 dated earlier, 6 later
1612.00593 (2016-12)
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
1803.04996 (2018-03)
Comparing Task Simplifications to Learn Closed-Loop Object Picking Using Deep Reinforcement Learning
1805.01026 (2018-05)
Computing CNN Loss and Gradients for Pose Estimation with Riemannian Geometry
and 11 more not named here
translation_from_matrix 14 other papers
8 dated earlier, 6 later
1612.00593 (2016-12)
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
1803.04996 (2018-03)
Comparing Task Simplifications to Learn Closed-Loop Object Picking Using Deep Reinforcement Learning
1805.01026 (2018-05)
Computing CNN Loss and Gradients for Pose Estimation with Riemannian Geometry
and 11 more not named here
translation_matrix 14 other papers
8 dated earlier, 6 later
1612.00593 (2016-12)
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
1803.04996 (2018-03)
Comparing Task Simplifications to Learn Closed-Loop Object Picking Using Deep Reinforcement Learning
1805.01026 (2018-05)
Computing CNN Loss and Gradients for Pose Estimation with Riemannian Geometry
and 11 more not named here
read_txt 2 other papers
2 later
2312.05391 (2023-12)
Loss Functions in the Era of Semantic Segmentation: A Survey and Outlook
2403.03532 (2024-03)
Extend Your Own Correspondences: Unsupervised Distant Point Cloud Registration by Progressive Distance Extension
Dy 1 other paper
1 later
2207.10047 (2022-07)
Densely Constrained Depth Estimator for Monocular 3D Object Detection
correspondenceProbabilityDistances 1 other paper
1 later
2207.10047 (2022-07)
Densely Constrained Depth Estimator for Monocular 3D Object Detection
get_file_list 1 other paper
1 later
2403.03532 (2024-03)
Extend Your Own Correspondences: Unsupervised Distant Point Cloud Registration by Progressive Distance Extension
pairwiseL2Dist 1 other paper
1 later
2207.10047 (2022-07)
Densely Constrained Depth Estimator for Monocular 3D Object Detection
sorted_alphanum 1 other paper
1 later
2403.03532 (2024-03)
Extend Your Own Correspondences: Unsupervised Distant Point Cloud Registration by Progressive Distance Extension

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Blind Perspective-n-Point (PnP) is the problem of estimating the position and orientation of a camera relative to a scene, given 2D image points and 3D scene points, without prior knowledge of the 2D-3D correspondences. Solving for pose and correspondences simultaneously is extremely challenging since the search space is very large. Fortunately it is a coupled problem: the pose can be found easily given the correspondences and vice versa. Existing approaches assume that noisy correspondences are provided, that a good pose prior is available, or that the problem size is small. We instead propose the first fully end-to-end trainable network for solving the blind PnP problem efficiently and globally, that is, without the need for pose priors. We make use of recent results in differentiating optimization problems to incorporate geometric model fitting into an end-to-end learning framework, including Sinkhorn, RANSAC and PnP algorithms. Our proposed approach significantly outperforms other methods on synthetic and real data.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2007.14628")
get_code_for_paper("2007.14628")
have("2007.14628")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2007.14628.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2007.14628.svg)](https://syntology.ai/paper/2007.14628)

Connect an agent — have() is free.